DISCOLOURING OF GRAPE
JUICE CONCENTRATE:
CAUSES AND POSSIBLE WAYS
OF INHIBITION
by
Matthys Johannes Loedolff A thesis submitted in partial fulfilment of the requirements for the
degree of
Master of Science in Engineering (Chemical Engineering)
Department of Process Engineering University of Stellenbosch
Supervisor: Prof. L. Lorenzen
DECLARATION
I hereby declare that the work done, the results obtained and the conclusions made are the work of my own. Wherever information was used and obtained from references it is so stated.
Signed on this ____ day of _________ 2004
SUMMARY
The grape juice concentrate (GJC) plant of the KWV at Robertson spent significant amounts of money on the decolourisation of grape juice concentrate. A chemically activated powdered activated carbon (PAC) purchased from Norit, namely CA1, was used as decolourisation product. Apart from the expenses involved, it contributed largely to the solid waste produced at this plant. A way was sought to minimise or prevent GJC discolourisation (and possibly solid waste) without increasing operating expenses.
Browning reactions in GJC are as old as the product itself. Numerous researchers have studied the origins of these reactions, the reactants and products involved, as well as the reaction kinetics of these reactions. From the work of these researchers four possible browning reaction pathways were identified, namely:
• enzymatic oxidative browning, • non-enzymatic oxidative browning,
• non-enzymatic browning (the Maillard reaction), and • caramelisation.
It was also identified that 5-hydroxymethylfurfural (HMF) are indicative of the browning potential of GJC. A method to analyse for HMF (quantitative and qualitative) was develop for the purposes of this study, namely positive electron-spray ionisation preceded by high-pressure liquid chromatography (HPLC) and followed by dual mass spectrometry. This method showed good repeatability and was used to analyse all samples generated during this study.
It was confirmed that the manufacturing process at this plant favours non-enzymatic browning reactions, since mild heat treatment deactivates enzymes. Further investigation indicated that the overruling browning reaction on this plant was non-enzymatic oxidative browning. It was shown that neither the presence, nor the absence of protein had any effect on the rate of formation of HMF. It was, however, confirmed that HMF formation could be attributed to high temperatures and prolonged exposure to these temperatures.
Other adsorption products were evaluated against the then current PAC (CA1), namely a steam activated PAC supplied by Norit, SA4, and a polymeric adsorbent, Polyclar V (polyvinylpolypyrrolidone/PVPP). Both SA4 and PVPP indicated superior HMF adsorption capacities. Replacing CA1 with SA4 could result in operating expenses savings and possible solid waste reduction. However, PVPP were too expensive to be considered an economically viable replacement for CA1.
Improved concentration technologies such as reverse osmosis (RO) membrane concentration followed by centrifugal evaporation (CE) or two-stage CE should be considered as possible replacement for the existing concentration technology (multi-stage falling film evaporator). This should decrease heat treatment/exposure by more than 90% and thus reduce browning significantly. An added advantage could be the reduction of solid waste, since less (if not no) decolourisation will be required.
Alternatively, juice should be stored with added sulphur dioxide (SO2),
since it was shown that this juice contained much lower HMF concentrations than diluted concentrate (stored for the same time). This should reduce heat exposure by up to 50% and thus minimise browning reactions.
OPSOMMING
Die druiwesapkonsentraat (DSK) aanleg van die KWV in Robertson het jaarliks aansienlike bedrae geld spandeer tydens die ontkleuringsproses van DSK. ‘n Chemies geaktiveerde verpoeierde koolstof (GVK) verkrygbaar van Norit, naamlik CA1, is gebruik as ontkleuringsproduk. Buiten die kostes verbonde aan hierdie produk het dit ook grootliks bygedra tot soliede afval by hierdie aanleg. Oplossings is gesoek om die verbruining/ontkleuring van DSK (en dalk ook soliede afval) te verminder (of selfs te voorkom) sonder om bedryfskostes te verhoog.
Verbruiningsreaksies in DSK bestaan al so lank soos DSK self. Verskeie navorsers het die oorsake, reaktante, produkte en reaksiekinetika van hierdie reaksies oor die jare heen bestudeer. Uit die werk van sommige van hierdie navorsers kon vier moontlike verbruiningsreaksieroetes geïdentifiseer word, naamlik:
• ensiematiese oksidatiewe verbruining, • nie-ensiematiese oksidatiewe verbruining,
• nie-ensiematiese verbruining (die Maillard-reaksie), en • karamelisering.
Daar was verder geïdentifiseer dat 5-hidroksiemetielfurfuraal (HMF) aanduidend is van die verbruiningspotensiaal van DSK. ‘n Analitiese metode (kwalitatief en kwantitatief) om vir HMF te analiseer is vir die doel van hierdie studie ontwikkel, naamlik positiewe elektronsproei ionisasie, voorafgegaan deur hoëdruk vloeistof chromatografie en gevolg deur dubbele massa spektrometrie. Hierdie analitiese metode het goeie herhaalbaarheid getoon en was deurgaans gebruik om monsters te analiseer gedurende hierdie studie.
Dit was bevestig dat die vervaardigingsproses by hierdie aanleg nie-ensiematiese verbruiningsreaksies begunstig, aangesien geredelike hittebehandeling ensieme deaktiveer. Verdere navorsing het getoon dat die oorheersende verbruiningsreaksies by hierdie aanleg nie-ensiematiese oksidatief van aard is. Resultate het getoon dat proteinstabiliteit geen invloed op die vormingstempo van HMF het nie. Dit was egter bevestig dat vorming van HMF direk verband hou met hoë temperature en lang blootstellingsperiodes aan hierdie temperature.
Ander adsorpsieprodukte was vergelyk met die huidige GVK (CA1), naamlik ‘n stoom geaktiveerde verpoeierde koolstof (Norit se SA4) en ‘n polimeriese adsorbant, Polyclar V (polivinielpolipirrolidoon/PVPP). Beide SA4 en PVPP het CA1 oortref wat betref HMF adsorpsie. Moontlike bedryfskostebesparings (en soliede afval verminderings) potensiaal bestaan indien CA1 vervang word met SA4. Die teenoorgestelde is egter waar vir PVPP wat bedryfskoste aangaan.
Instede van die huidige verdampinstegnologie, naamlik vallende-filmverdamping, hoort verbeterde konsentrasietegnologieë soos tru-osmose membraankonsentrasie gevolg deur sentrifugale verdamping, of, alternatiewelik, twee-stadium sentrifugale verdamping, orrweeg te word. Op hierdie wyse behoort hittebehandeling (en dus verbruining) met sowat 90% verminder te word. ‘n Moonlike addisionele voordeel is die vermindering van soliede afval aangesien minder ontkleuring nodig sal wees.
Indien die verbeterde tegnologieë te duur is moet daar gekyk word daarna om die ongekonsentreerde sap met addisionele swaweldioksied (SO2) te
stoor, aangesien veel laer HMF konsentrasies in sulke sap waargeneem is as in verdunde direkte konsentraat wat vir dieselfde typerk gestoor is.
Hittebehandeling sal op hierdie wyse met tot 50% verminder word (en dus verbruining ook).
TABLE OF CONTENTS
Declaration ... ii
Summary... iii
Opsomming ... v
Table of Contents ... viii
List of figures ... xii
List of tables ... xvi
Acknowledgments... xvii
1 Introduction... 1
1.1 Background ... 1
1.1.1 Concentration by freezing... 1
1.1.2 Concentration by RO ... 2
1.1.3 Concentration by distillation processes ... 2
1.1.4 Application of GJC ... 2
1.2 Problems experienced during GJC manufacturing ... 3
1.2.1 Tartrate instability ... 3
1.2.2 Sugar crystallisation ... 3
1.2.3 Fermentation... 3
1.2.4 Browning of the concentrate during storage ... 4
1.2.5 Oxidative browning... 4 1.2.6 Non-enzymatic browning ... 4 1.2.7 Caramelisation... 4 1.2.8 In general ... 5 1.3 Scope of work... 5 1.4 Objectives of study... 6
2 Background and literature study ... 8
2.1 Background on the conventional process... 8
2.1.2 Stage 2a - Direct concentration and storage... 9
2.1.3 Stage 2b - Dilution ... 9
2.1.4 Stage 2 - Desulphurisation process... 9
2.1.5 Stage 3 - 1st Concentration... 10
2.1.6 Stage 4 - Protein stabilisation (and decolourisation)... 10
2.1.7 Stage 5 - Filtration... 10
2.1.8 Stage 6 - Cooling... 11
2.1.9 Stage 7 - Tartrate Stabilisation... 12
2.1.10 Stage 8 - Filtration... 12
2.1.11 Stage 9 - 2nd Concentration ... 12
2.1.12 Stage 10 - Blending... 12
2.1.13 Stage 11 - Pasteurisation ... 12
2.1.14 Stage 12 - Drum filling/storage ... 13
2.2 Literature study ... 13
2.2.1 The browning problem in juice and methods of dealing with it... 13
2.2.2 The chemistry of browning reactions ... 20
2.2.3 Applications and limitations of adsorption products ... 28
2.2.4 Background on products chosen for this study ... 36
2.3 Summary ... 42
3 Analysis: Method development ... 44
3.1 Method development ... 44
3.1.1 Qualitative and quantitative classification of the phenolics in GJC ... 44
3.1.2 Background on Methods Used ... 45
3.1.3 Trials... 51
3.1.4 Inadequate results... 56
3.1.5 Near successful results ... 70
3.1.6 Selected method of analysis: +ESI-MS-MS ... 77
4 Experimental... 86
4.1 Determining the effect of the conventional process on the concentration of HMF... 86
4.1.1 The effect of the conventional process by analysis of samples taken after different stages of production... 86
4.1.2 The effect of heat induced over a period of time on two grape juice samples, one protein and tatrate stable, and one not protein and tartrate stable... 88
4.2 Comparison of the effect of the adsorption products on the concentration of HMF... 89
4.2.1 Determining optimum conditions for the products ... 89
4.2.2 The product profiles ... 90
4.3 General observations during experimental work ... 91
5 Results and Discussion... 92
5.1 Background and literature study... 92
5.1.1 The conventional process... 92
5.1.2 Decolourisation products ... 93
5.1.3 Chemistry of discolouring (browning) reactions ... 94
5.2 Development of method of analysis to effectively qualify and quantify HMF... 95
5.3 Adsorption products... 95
5.4 Investigating the effect of the conventional process on the relative concentration of HMF... 96
5.5 Investigating the effect of heat on HMF concentration for both a protein (and tartrate) stable and unstable juice... 98
5.5.1 Rate of HMF formation based on analysis results from heat experiment ... 100
5.6 Product Profiles ... 103
5.7 Investigating the effect of three adsorption products on the relative concentration of HMF... 105
5.7.1 The effect of contact time and contact temperature on the
relative HMF concentration ... 105
5.8 Summary ... 109
6 Evaluation of the conventional process ... 111
6.1 Possible alterations to the conventional process ... 111
6.1.1 Protein stabilisation... 111
6.2 Decolourisation product... 113
6.3 Alternative concentration technologies ... 113
6.4 Comparing running costs of the conventional process and the conventional process with suggested changes ... 114
6.4.1 Protein stabilisation... 114
6.4.2 Decolourisation product... 115
7 Conclusions and recommendations... 116
7.1 Background and literature study... 116
7.2 Analysis: Method development ... 117
7.3 Experimental ... 118
7.4 Evaluation of the conventional process... 119
8 References... 121 9 Nomenclature... 127 9.1 General nomenclature ... 127 9.2 Sample nomenclature... 127 9.2.1 General samples: ... 127 9.2.2 Robertson sampling: ... 128 9.2.3 Heat experiments:... 128 9.2.4 Optimum conditions: ... 128 9.2.5 Product Profiles:... 128 10 Appendices ... 130
Appendix A – Technical data: Adsorption products... 131
Appendix B – Typical calculations... 142
LIST OF FIGURES
Number Page
Figure 2-1 Simplified process flow diagram of the conventional process .11 Figure 2-2 A few antioxidants/phenolics participating in oxidative
browning reactions...22
Figure 2-3 Enzymatic browning of catechin, the formation of a brown polymer...23
Figure 2-4 Non-enzymatic oxidative browning of a leucocyanidin ...23
Figure 2-5 Formation of Schiff base and Amadori compounds ...24
Figure 2-6 The Maillard reaction pathway to HMF and melanoidins...25
Figure 2-7 Acid catalysed sugar dehydration (The formation of HMF) ....27
Figure 2-8 Three-dimensional graphite lattice pore structure of a typical activated carbon particle ...38
Figure 2-9 Mechanism 1 - PVPP with same binding sites as proteins ...41
Figure 2-10 Mechanism 2 - PVPP with different binding sites than proteins ...41
Figure 3-1 MALDI-TOF...46
Figure 3-2 Time of Flight (t) in the field-free region (D = distance travelled by ion, m/z = mass to charge ratio, V = guide voltage)...47
Figure 3-3 Liquid Chromatography, a simplified illustration...48
Figure 3-4 Simplified illustration of LC/MS/MS...50
Figure 3-5 Mass Spectra of analyte and matrix (MALDI-TOF) ...57
Figure 3-6 Mass Spectra with smaller m/z-range...57
Figure 3-7 Chromatogram obtained from pooled sample...58
Figure 3-8 Chromatogram obtained from a Grape Seed Extract ...59
Figure 3-9 Technically incorrect comparison of the two profiles ...59
Figure 3-10 Chromatograph of old GJC sample ...59
Figure 3-12 Comparison of profiles ...60
Figure 3-13 Mass spectra of Quercetin molecular ion (A), and fragmentation spectra (B), ESI ...61
Figure 3-14 Mass spectra of Catechin molecular ion (A), and fragmentation spectra (B), ESI ...61
Figure 3-15 Mass spectra of Epicatechin molecular ion (A), and fragmentation spectra (B), ESI ...62
Figure 3-16 Mass spectra of Caffeic Acid molecular ion (A), and fragmentation spectra (B), ESI ...62
Figure 3-17 Mass spectra of Vanillic Acid molecular ion (A), and fragmentation spectra (B), ESI ...63
Figure 3-18 Mass spectra of Gallic Acid molecular ion (A), and fragmentation spectra (B), ESI ...63
Figure 3-19 Mass spectra of Resveratrol molecular ion (A), and fragmentation spectra (B), ESI ...64
Figure 3-20 Mass spectrum of 5-Hydroxymethylfurfural, ESI ...65
Figure 3-21 Mass chromatogram of quercetin, ESI ...66
Figure 3-22 Mass chromatogram of catechin, ESI...67
Figure 3-23 Mass chromatogram of epicatechin, ESI...67
Figure 3-24 Mass chromatogram of catechin/epicatechin mixture, ESI ....68
Figure 3-25 Mass chromatogram of caffeic acid, ESI ...68
Figure 3-26 Mass chromatogram of gallic acid, ESI ...69
Figure 3-27 Mass chromatogram of vanillic acid, ESI ...69
Figure 3-28 Mass chromatogram of resveratrol, ESI...70
Figure 3-29 Mass spectra of Quercetin molecular ion (A), and fragmentation spectra (B), APcI...71
Figure 3-30 Mass spectra of Catechin molecular ion (A), and fragmentation spectra (B), APcI...71
Figure 3-31 Mass spectra of Caffeic Acid molecular ion (A), and fragmentation spectra (B), APcI...72
Figure 3-32 Mass spectra of Gallic Acid molecular ion (A), and
fragmentation spectra (B), APcI...72
Figure 3-33 Mass spectra of Vanillic Acid molecular ion (A), and fragmentation spectra (B), APcI...73
Figure 3-34 Mass spectra of Resveratrol molecular ion (A), and fragmentation spectra (B), APcI...73
Figure 3-35 Mass spectra of HMF molecular ion (A), and fragmentation spectra (B), APcI...74
Figure 3-36 Compounds injected separately ...75
Figure 3-37 Compounds injected separately ...75
Figure 3-38 Juice sample injected for detection, ApcI...76
Figure 3-39 Juice sample injected for detection, ApcI...76
Figure 3-40 Simplified illustration of LC-ESI-MS-MS...79
Figure 3-41 Calibration curve 1...80
Figure 3-42 Calibration curve 2...81
Figure 3-43 Calibration curve 3...81
Figure 3-44 Calibration curve 4...82
Figure 5-1 Effect of conventional process and storage at different conditions on HMF concentration during the GJC manufacturing process ...97
Figure 5-2 Comparison of protein (and tartrate) stable/unstable juice at 100oC and 20 minute intervals between sampling...99
Figure 5-3 First order reaction rate of HMF formation at 100oC ...102
Figure 5-4 Second order reaction rate of HMF formation at 100oC...102
Figure 5-5 Effect of CA Carbon dosage dosage on both stable and unstable juice at 20oC and six hours contacat time ...103
Figure 5-6 Effect of SA carbon dosage on both stable and unstable juice at 20oC and six hours contact time ...103
Figure 5-7 Effect of PVPP dosage on 5-HMF in both stable and unstable juice at 20oC and six hours contact time ...104
Figure 5-8 Effect of temperature on CA Carbon at various contact times and dosage of 4 g/L...108 Figure 5-9 Effect of temperature on SA carbon at various contact times and dosage of 4 g/L...108 Figure 5-10 Effect of temperature on PVPP at various contact times and
dosage of 0.5 g/L...109 Figure A-1 Chemical structure of Polyclar® stabilizers...140
LIST OF TABLES
Table Page
Table 2-1 A summary of the preferred environments of the browning
reactions ...27
Table 2-2 Examples of grape juice phenolics ...28
Table 2-3 General specifications of NORIT CA1...39
Table 2-4 General specifications of NORIT SA4 ...40
Table 2-5 General specifications of Polyclar V ...42
Table 3-1 Molecular- and fragment ion combinations, ESI ...65
Table 3-2 Molecular- and fragment ion combinations, APcI ...74
Table 4-1 Product dosages ...90
Table 4-2 Conditions for determining optimum product conditions ...90
Table 4-3 Dosage masses for CA1, SA4, and PVPP ...91
Table 5-1 A summary of the preferred browning reaction environments ..94
Table 5-2 Cummulative sorption area of CA1 and SA4...96
Table 5-3 Reaction rate determination ...101
Table 5-4 Comparison of HMF removal efficiency...105
Table 6-1 Operating cost comparison o different adsorption products ....115
ACKNOWLEDGMENTS
To everybody that contributed to this thesis, thank you. I want to thank Prof. Leon Lorenzen, my study leader for guidance during this study. Also, to Dr. Thinus van der Merwe who worked long hours on the analysis that was essential to complete this study. Mr Niel Rademeyer, the technical officer on the plant this study was based on, a huge thank you for assisting me even when you didn't have the time. To Jannie Barnard, Anton Cordier, and Howard Koopman who helped constructing the experimental setup. Also thanks to Vincent, Charles and James for all the assistance.
To my parents: "Pa, Ma, dankie vir alles! Ek weet julle het baie opgeoffer in die tyd wat ek hier op Stellenbocsh deurgebring het (en voor dit ook!!). Ek is baie lief vir julle!"
To my wife: "Marli, my skat, dankie vir die ondersteuning en vir die tyd wat jy broodwinner was in die huis. Ek is oneindig lief vir jou!"
To all that showed an interest during this few years, my friends, my younger sister, and my in-laws, thank you for the support.
Special thanks to my financial sponsors, Winetech and NRF, who funded this project. It is much appreciated.
And lastly, but the most, to the Lord of lords and the King of kings, Jesus: Thank You for your strength and the abilities You have given me. I acknowledge You as Lord of my life.
1 I
NTRODUCTIONThe process where grape juice is boiled to what is known in South Africa as "Moskonfyt" (jam), is a very old method of grape juice concentrate (GJC) manufacturing. It was the first method in which grape must or "mos" was concentrated. The original process involved the boiling of the grape juice in open pots at atmospheric pressure. This caused the boiling point to increase as the water evaporated. Due to this increase in temperature, thermal degradation of the sugars and other compounds occurred, causing the jam to lose most of its grape flavour and to obtain a "boiled" taste. "Moskonfyt" was not an acceptable GJC and other methods of concentration were developed.
1.1 BACKGROUND
The three major grape juice concentration technologies used are: • Concentration by freezing;
• Concentration by reverse osmosis (RO); • Concentration by distillation processes. 1.1.1 Concentration by freezing
Concentration by freezing involves the exposure of the grape juice to intense cold (down to -14oC). The water fraction of the juice freezes under these conditions. The small ice crystals can then be removed by a specialized centrifuge able to separate the ice crystals from the concentrate while washing the ice with clean water to remove the remaining grape juice. What remains is the concentrated juice. This process is more energy effective than any distillation process. However, operating and capital expenses do not warrant its use for GJC manufacturing.
1.1.2 Concentration by RO
Reverse osmosis is often used to adjust sugar content in grape juice, to adjust alcohol and volatile acid levels in wines, or to concentrate wines. The juice is pumped to flow across the one side of a selectively permeable membrane. The water then selectively permeates the RO-membrane leaving a concentrated grape juice on the concentrate-side of the membrane. Applying pressure to force more water through the membrane can further increase the effect of concentration. However, there is an optimum operating point determined by product yield and operating cost (input energy, etc.). Pressures as high as 80bar(g) is required to obtain a GJC with a maximum sugar concentration of 50% (wt/v). Thus, the maximum attainable sugar concentration is approximately two thirds of the required concentration.
1.1.3 Concentration by distillation processes
The most common method of grape juice concentration is multiple effect evaporation (MEE). The plant, on which this study was conducted, made use of MEE. Juice is boiled (at approximately 70oC maximum) under vacuum, causing evaporation of water. The vacuum draws the vapour off and a GJC remains as product.
1.1.4 Application of GJC
GJC is used in the following applications: • The sweetening of table wines;
• Some countries that do not commercially produce grapes, reconstitute the GJC, and then inoculate it with yeast to produce their own wines;
• The base for fruit juices and cooled drinks (also carbonated); • Baby foods, ice creams, and yoghurt.
1.2 PROBLEMS EXPERIENCED DURING GJC MANUFACTURING
The major single problems currently experienced during concentrate manufacturing are:
• Tartrate instability; • Sugar crystallisation; • Fermentation;
• Browning of the concentrate during storage. 1.2.1 Tartrate instability
Tartrates are salts that form when minerals such as Potassium and Calcium combine with natural acids (tartaric acid) in the juice. During the concentration process, water is removed, causing the tartrate concentration to increase. This increase in tartrate concentration causes tartrates to visibly precipitate. The murkiness is unacceptable to consumers and must thus be removed. Tartrate stabilisation alleviates this problem.
1.2.2 Sugar crystallisation
The crystallisation of sugar at the bottom of the tanks is a problem that already occurs when the concentrate is in cold storage (might in some instances be confused with tartrate precipitation). Sugar crystallising in the shipping containers pose a major problem. Receiving clients deem such juice as unacceptable. Tests and practice have proved that the addition of citric acid during container filling alleviates this problem, almost entirely. 1.2.3 Fermentation
Fermentation is can be prevented by good hygiene. All apparatus should be thoroughly cleaned and sterilised before and after use.
1.2.4 Browning of the concentrate during storage
During long periods of storage, GJC tend to degrade visibly by discolouring (also referred to as browning). Colour changes from a golden yellow to dark brown. Three possible browning pathways exist. They are:
• Oxidative browning (including enzymatic and non-enzymatic); • Non-enzymatic oxidative browning;
• Caramelisation. 1.2.5 Oxidative browning
During storage of fruit juice or fruit juice concentrate some of the phenolic compounds present in the juice tend to react via an oxidation pathway (either catalysed by enzymes or an acidic environment) to form larger phenolic molecules or polyphenolics. After some time the colour turns to dark brown. This phenomenon can be seen in some bottled apple and grape juices if the colour of a rather new batch is compared with the colour of a rather old batch.
1.2.6 Non-enzymatic browning
The Maillard reaction in foodstuffs is very complex and, even today, is not fully understood. This study focuses only on two of the pathways within the entire Maillard reaction, namely the ones involving the formation of 5-hydroxymethylfurfural (HMF).
1.2.7 Caramelisation
Caramelisation does not form part of this study, since it involves very high temperatures applied for significant periods of time with little to no water. Such high temperatures are not reached during the grape juice concentration process we are focusing on.
1.2.8 In general
There are a number of products that can remove the brown colour from the juice. However, the most common method of decolourising is by contacting the juice with activated carbon for a few hours, followed by filtration to remove any suspended solids. This method is sufficient for dealing with the symptoms of the browning reaction. However, the efficiency of removing the precursors of the browning reactions remains a challenge.
1.3 SCOPE OF WORK
This study will focus on the cause and possible elimination of browning reactions. This study involved one grape juice concentration plant where browning reactions are a problem. This plant is situated in Robertson, Western Cape, South Africa, and is the property of the KWV (this plant has stopped operation during the course of this study). A detailed description of the process used in manufacturing the concentrate, is given in Chapter 2 (refer to Figure 2-1).
The browning of the concentrate was the direct implication of added expenses on this specific plant. After long periods of storage the GJC required decolourisation. When fresh juice is received at the plant it is directly concentrated to 70+ oBalling. This juice is then stored at 10oC for a few months. During this storage period, the colour changes from a golden yellow to a dark brown, which is unacceptable. The concentrate has to be reworked. This involves the dilution of the concentrate, decolourisation of the juice, and re-concentration. Needless to say, it is a costly operation.
The possibility exists for the manufacturing process to be altered slightly to possibly eliminate (or scale down) the decolourisation process, or at least
reduce solid wastes generated and the raw materials required during this process.
1.4 OBJECTIVES OF STUDY
The objectives of this study can be summarised as follows:
1. Background and literature study. This includes assessing the conventional GJC manufacturing process searching for references focused on decolourisation products, existing decolourisation procedures, the chemistry of discolouring (browning) reactions, adsorption products, etc.
2. Development of a method of analysis to effectively qualify
and quantify 5-hydroxymethylfurfural (HMF). It involves
the qualification and quantification of HMF, which is a product of the Maillard reaction and/or non-enzymatic oxidation. HMF is indicative of the browning potential of GJC. HMF is a precursor to browning, since it only needs to react with an amine to produce a melanoidin (Dutson & Orcutt, 1984). Melanoidins are visible as a brown colour in juice and juice concentrate.
3. Investigating the effect of the conventional process on the
concentration of HMF. This involves the sampling of the
juice after each step of the conventional concentration process, qualifying and quantifying HMF, and drawing conclusions on the results obtained. The aim is to determine which pathway of browning is the most predominant.
4. Investigating the effect of three adsorption products on the
concentration of HMF. The aim of this objective is to
compare the effects of a few products, such as polyvinylpolypyrrolidone (PVPP), a steam activated
carbon, and a chemically activated carbon, on the concentration of HMF under various process conditions. This objective will include three different experimental procedures, specifically chosen to determine the effect of temperature, time, dosage volume, and protein stability on the HMF removal efficiency of the three adsorption products. One experimental procedure will be aimed at determining the effect temperature has on the constitution of both a protein stable and unstable juice.
5. Suggesting a possible change in the conventional process to
minimise treatment, minimise waste production, and ensure a longer storage life of the product. This objective is thus
focused towards the optimisation of the existing process. 6. Comparing the running costs of the conventional process
and the suggested process. This objective will indicate the
viability in the change in process and will take into account cost and dosage volumes of the adsorption products.
2 B
ACKGROUND AND LITERATURE STUDYAs mentioned previously this study is focused on one specific grape juice concentration plant using one specific process to concentrate grape juice. This process will be termed: "The conventional process."
2.1 BACKGROUND ON THE CONVENTIONAL PROCESS
The conventional process consists of the following main stages. (A simplified process flow diagram is given in Figure 2-1)
2.1.1 Stage 1 - Harvesting/crushing
The grapes are harvested on the wine farms and taken to the cellar where the crushing is performed. The quality of the grapes is usually determined by the sugar content thereof. By determining the sugar content, in degrees Brix (oBrix), the winemaker is able to determine whether the grapes are fit for making wine and/or juice concentrate. The minimum sugar content of the juice allowed by the concentrate manufacturer is approximately 16.5
o
Brix. The temperature at which the grapes should be delivered to the cellar is 12 to 20oC. The grapes are crushed and the juice is either stored at the cellar or delivered directly to the concentrate manufacturer. SO2 is
added to the juice during storage. When the juice is stored at the cellar for longer than a weekend (3 days), approximately 1200 to 1400 mg/L SO2 is
added to the juice. This juice will be called: SO2-juice. After a few
weeks/months this stock will then be delivered to the juice concentration plant, depending on the need of the clients. If the juice is taken to the concentration plant on the day of harvest, the allowed SO2-content is
between 150 and 200 mg/L. If the harvest came in on a Friday and has to remain at the cellar of harvest during the weekend, the allowed content of SO2 is 400 mg/L. This juice will be called: fresh juice.
SO2-juice and fresh juice are handled differently in the following stage.
SO2-juice simply continues from Stage 2 onwards. Fresh juice is
incorporated into the manufacturing process by stages 2a and 2b. 2.1.2 Stage 2a - Direct concentration and storage
Fresh juice is directly concentrated to 70+oBalling without undergoing any protein or tartrate stabilisation, etc. From here it is pumped to a storage facility where it is stored for up to nine months at 10oC.
2.1.3 Stage 2b - Dilution
Depending on the requirements of the client, the stored concentrate is then diluted to e.g. 20oBalling by adding water. From here on forward the process for both fresh and SO2-juice is the same (stage 2).
2.1.4 Stage 2 - Desulphurisation process
The diluted fresh juice/SO2-juice is preheated by the desulphited product
stream of the stripper column to a temperature of 86oC. It is then pumped to the stripper column and fed to the top plate of the column. Gas from the outlet at the top of the desulphurisation (stripping) column enters at the bottom of lime tanks (simplified) where suplhur dioxide is removed from the air. The SO2 is removed from the juice by vapour rising to the top of
the column while in contact with the falling juice. The bottoms, or residue is pumped to a reboiler where steam heats it to a temperature of 90oC. The generated vapour is fed to the bottom tray of the column. The product stream (desulphited juice) is collected from the reboiler. The vapour from the stripping column is fed to a condenser, which is cooled by cooling water. The SO2 is drawn of to lime tanks while the condensate is returned
to the column to prevent juice concentration during this stage of the process. The lime tanks are kept under vacuum. The vacuum draws the SO2 from the distillate and bubbles it through the lime solutions in the lime
tanks. A pH-meter constantly measures pH in the lime tanks giving an indication of when the lime is spent.
The product stream then preheats the feed stream and is cooled down to a temperature of 55oC. From here it goes to the first concentration.
2.1.5 Stage 3 - 1st Concentration
Concentration is established by a multiple effect evaporator. This process uses a double-effect evaporator. The first effect is operated at a vacuum of -40 kPa(g) and the second effect at -85 kPa(g). The steam flows at 9 ton/hr at a temperature of 105oC, while the juice exits at 18 000 L/hr. The concentration increases from 17 to 35 oBalling and the exit temperature is 55oC.
2.1.6 Stage 4 - Protein stabilisation (and decolourisation)
Protein stabilisation takes place over a period of seven hours, at a temperature of approximately 55oC. It is done in four tanks, two 600 hL and two 400 hL tanks. It involves the addition of calcium-bentonite and activated carbon. The bentonite is added in doses of 140 g/hL, but in the form of a 10% (mass/volume) mixture in water. The activated carbon is added to the juice in doses of 400 g/hL. A natural acid is added to the juice when the acidity is too low, while potassium carbonate is added when the acidity is too high. The solid wastes are discarded after filtration.
2.1.7 Stage 5 - Filtration
The filter used on this plant is a stainless steel sheet filter, with a mesh size of 9 micron, currently handling a juice flow of 20000 L/hr. The filtering medium is diatomaceous earth and one kilogram can filter 1250 litres of juice.
FIGURE 2-1SIMPLIFIED PROCESS FLOW DIAGRAM OF THE CONVENTIONAL PROCESS
2.1.8 Stage 6 - Cooling
Cooling on this plant is performed by cooling water and plate exchangers. The cooling water is cooled in two stages in the cooling plant. The first stage is pre-cooling where the water is cooled to as low as 0oC, and the second stage is crash-cooling where the water is taken to -4oC. This is done at a water flow rate of 18000 L/hr.
Stage 1- delivery Fresh Juice or SO2 -Juice 1st 2nd Stage 2a -direct concentration to storage, skip Stages 4 - 8 Storage at 10 degrees Celsius Break down tanks - water added SO2-Juice to Stage 2, and Fresh Juice to Stage 2a To Stage 2 Stage 2b -Breaking Down Stage 2 -Desulphiting Process Condensate volume water back to process, this stream to lime tanks Stage 3 - 1st Concentration Protein Stabilisation and decolourisation Stage 4 Stage 5 - Filtration Cooling plant Stage 6 - Cooling
Stage 7 - Tartrate Stabilisation Stage 8 - Filtration Stage 9 - 2nd Concentration Stages 10, 11, and 12 Blending, Pasteurisation, Drum Filling
2.1.9 Stage 7 - Tartrate Stabilisation
The juice is sent to the cold stabilisation tanks where it is cooled to a temperature of -4oC. Natural tartrates are recovered from cooling the juice. The juice is then dosed with a 55% (solids) tartrate solution at 30 g/L juice. The process is continuous, flowing through six tanks, each with a volume of 12000 L, at a flow rate of 16000 L/hr. Filtration then follows.
2.1.10 Stage 8 - Filtration
The same filter is used as in stage 5. The pH barely changes during tartrate stabilisation from 3.4 to 3.5, and should in fact remain constant since tartrates are merely naturally occurring salts.
2.1.11 Stage 9 - 2nd Concentration
The second concentration is basically the same as the first concentration with the main differences that the steam temperature is now 85oC, the exit flow rate of the juice is 9000 L/hr, and its exit temperature is 10oC.
2.1.12 Stage 10 - Blending
The juice concentrate is blended to the requirements of the clients. Some clients prefer lower sugar content, so the concentrate will be diluted to the
o
Brix specified by the client. This is done in blending tanks (of which there are three), each with a volume of 15000 L.
2.1.13 Stage 11 - Pasteurisation
The juice concentrate is standardised at the required oBrix as mentioned above. After standardisation the concentrate gets pasteurised. The concentrate is heated to a temperature of 86oC and kept there for a few seconds. It is then cooled down again to 5oC. The pasteurisation plant is able to provide 85% of the necessary heat by means of counter-current (regenerative) heating. This then implies that only 15% of the heat is to be
supplied by the steam kettles. The pasteurisation plant consists of three units, namely:
• The generative heating section; • The hot water heating section: • The glycol cooling section.
The regenerative heating section is maintained by a counter current flow of concentrate from the hot water section with concentrate fed to the pasteurisation plant. The hot water section is heated by the hot water system of the plant, which makes use of steam to heat the water. The glycol cooling section cools the concentrate by counter current flow of the cooled concentrate from the regenerative heating section with cold glycol. 2.1.14 Stage 12 - Drum filling/storage
The product is then put into drums at a volumetric flow rate of 3750 L/hr. These drums are then stored at ambient temperature after filling until delivery or collection. Citric acid is usually added to increase the acid content to about 2 g/L to prevent sugar crystallisation (if required).
2.2 LITERATURE STUDY
2.2.1 The browning problem in juice and methods of dealing with it Two major methods of dealing with the browning problem in juices, juice concentrates, and wines are found in literature. They are:
• The prevention of the browning reactions by treating the clear juice in some way as to prevent browning;
• Curing of the juice by removing the brown colour formed during storage.
2.2.1.1 Prevention
The following summarised references contain some information on preventing browning in wine and/or juice.
Cantarelli et al. (1971) investigated the addition of formaldehyde for the prevention of browning of white wines. The addition of formaldehyde or hexamethylenetetramine (HMTA) to grape musts was marked by reduced phenolics in wines. The formaldehyde selectively precipitated the non-tannin flavonoid fraction and effectively slowed oxidative browning (“maderization”) in white wines. Similar detanninising effects were obtained with the insoluble PVP (polyvinylpirrolidine) also tested during this investigation. However, in their case the tannic flavonoid fraction is selectively eliminated, resulting in a decolourisation of the wines and a lower stability to maderization. The PVP was added after fermentation. Other detanninising agents that was also tested and compared with formaldehyde were: gelatin and casein, added to the must before fermentation. The concentration of formaldehyde residues appears to be inversely related to the flavonoid contents of wines. The excess of formaldehyde added to the must reacts with the sulphur dioxide, resulting in high residues of SO2 and formaldehyde in the wines. Although this
method of adding formaldehyde to the musts seemed effective, the residual formaldehyde is still a problem. Due to the interaction of the sulphur dioxide, there remain some difficulties in determining the amount of formaldehyde or HMTA, which has to be added to the must. Furthermore, to determine the exact dosage of formaldehyde or HMTA, one has to have a very good indication of the amount of non-tannin flavonoids present in each batch of must. A certain amount of formaldehyde reacts only with a certain amount of the non-tannin flavonoids. The rest is residual and has to be removed in some way. Formaldehyde is a carcinogen and residual formaldehyde remains a problem.
Kelly and Finkle (1969) investigated the action of a ring-cleaving enzyme in preventing oxidative darkening of juices. Catechol-type compounds such as chlorogenic acid are the natural substrates for polyphenol oxidase, a natural occurring enzyme in fruit juices. The dark pigments in juices are produced when catechol type compounds, i.e. dihydroxyaromatic compounds, present in fruit or juice, are oxidized in air through the action of catechol oxidase (or polyphenol oxidase) to polymeric quinones. The first step in this reaction involves the formation of an o-quinone. The quinone is then polymerised to dark pigments during further oxidation. Modifying the phenolic substrates in an irreversible manner so that they can no longer participate in the darkening reaction, can prevent the above reaction. The following three methods were used to alter the substrates: (1) O-methylation of catechol compounds; (2) an alkaline buffer treatment of cut fruit; and (3) oxidative ring opening of catechol compounds. Since this investigation was done on apples and apple juice, we will only discuss methods (1) and (3). Method (1) involves an enzymatic O-methylation of the catechols by treatment with the enzyme O-methyltransferase and a methyl donor such as S-adenosylmethionine. Since the O-methylation reaction is irreversible and the methylated products cannot be oxidized by catechol oxidase, oxidative darkening is then eliminated. Method (3) involves the oxidative ring opening of the catechol compounds. The enzyme used to accomplish this in apple juice was the ring-cleaving oxygenase, protocatechuate 3,4-dioxygenase (PC-ase). This enzyme converts protocatechuic acid to an open chain acid, beta-carboxymuconic acid. In a similar manner, it attacks other catechol acids, cleaving the aromatic ring and thus, preventing browning. These methods will not be given further attention in this study.
Peterson and Caputi (1967) studied the effects of Na+ exchange, H+ exchange, and H+ exchange followed by OH- exchange on browning of four dry white wines and compared them in the presence- and absence of
oxygen. Wines browned more when oxygenated than when stripped with N2, and ion-exchanged wines browned less than the untreated wines. H+
exchange was more effective than Na+ exchange in inhibiting the browning of all wines tested. H+ exchange followed by OH- exchange practically eliminated browning, but wines treated in this manner were water-white and contained little wine character. Evidence is also presented in this report that two distinct types of browning may occur simultaneously in wine, oxidative and non-oxidative. Oxidative involves the polymerisation of phenols and the oxidation of these polyphenols, whereas, non-oxidative possibly involves nitrogenous compounds such as amino acids condensing with carbonyl compounds. Caputi et al. (1969) further investigated residual PVP in wine and possible hazardous effects thereof.
Investigations done by Flores et al. (1988) on three vintages indicated that White Riesling juice processed without SO2 and clarified by ultra-filtration
(UF) tended to develop sediments on storage. This study analysed the possible effect of oxidation (processing with and without SO2) and pre-UF
treatment of juices with enzymes and fining agents on juice flux, colour (browning), composition, and stability. White Riesling juice was ultra-filtered with a Romicon Lab-5 pilot-scale hollow-fibre unit, operated in a batch mode, with membrane of nominal molecular weight cut-off (MWCO) of 10000 Daltons. Grapes were processed with and without SO2,
and the effects of treatment of settled press juice with Rohapect VR Super (VRS, mainly pectinase and protease) and of fining with bentonite, gelatine, and silica sol before UF were investigated. Juice parameters evaluated included total protein, pectin, phenol, colour (A420 nm), and
stability to heat/cold testing. Pre-UF treatment with enzymes and fining, increased flux. Sediments were found to contain large amounts of proteins and phenolics and trace amounts of pectin and neutral polysaccharides. Sediment formation and instability to heat testing of UF permeates processed without SO2 were prevented with pre-fining. Up to 99% of
protein, 90% of pectin, 84% of colour, and low variable phenolics were retained by the 10000-dalton-MWCO membrane. However, pre-UF treatment can increase protein and pectin in permeates. During UF, there is a significant increase in the soluble protein and water-soluble pectin passing through the membrane with increasing volume concentration ratio (VCR, process time). It is concluded that it is not only the quantity, but also the nature/state of compounds such as proteins, phenolics, and pectins and their interactions that results in instability. At present research is done on refining ultra-filtration techniques to such an extent that it becomes commercially acceptable (Borneman et al., 2001).
In a study done by Panagiotakopoulou and Morris (1991) juices from two different white grape cultivars, Aurore and Cayuga, received 13 different wine treatments to determine the effects on browning in wine. The additives used separately and in combination, were: ascorbic acid, hypo-phosphorous acid, thiodipropionic acid, Trolox-C, stannous chloride, Sporix, and SO2. All chemicals, accept SO2, added to wine were made
from nitrogen-sparged juice. One batch of the wine received no treatment. All treatments were stored at 20oC and 37oC for nine months and exposed to air after three and six months storage to accelerate browning. The use of SO2 during bottling resulted in the least browning, and the use of SO2
during crushing resulted in the most browning. Wine treated with the other chemicals had less browning than the wine produced from the nitrogen-sparged juice but had more browning than the untreated wine. The addition of ascorbic acid combinations reduced browning to a greater extent than any of the other chemicals or combinations.
Lee et al. (1990), Oszmianski & Lee (1990), Lee et al. (1985), McLellan et al. (1995), Lee & Whitaker (1994), and Kime & Lee (1987) investigated various aspects of the inhibitory effect of honey on polyphenol oxidase. They specifically gave attention to the possible reduction of SO2 in juice
and wine by using honey. From their results it appears that honey proved to be an efficient inhibitor of oxidative browning.
2.2.1.2 Cure
This section contains the summary of a few references containing information on the treatment of brown juice and/or wine.
Brú et al. (1995), treated different samples of dry sherry wine with several adsorbent compounds at different dosages. The evolution of the wine over a period of two years was then recorded. The following adsorbents were used: four types of activated carbon, two types of casein (Potassium Caseinate), two types of PVPP (Polyclar VT and Divergan W), and three adsorbent resins (XAD-761 Duolite, S-861 Duolite, and Micron 96). In the same way two control samples were prepared; one treated with Sodium Bentonite and the other without. The adsorbents were added to the wine in three dosages; 100, 500, and 2500 mg/L. It was concluded that, of all the adsorbents tested, the soluble casein and caseinates gave the best results in terms of browning removal. This was the case for low, medium, and high dosages of the compounds.
John Daumé (2001) gives some information in his website on a few fining agents used before bottling. Casein: Potassium Caseinate can improve both flavour and colour in slightly oxidized wines. Polyclar/PVPP: Microscopic, insoluble nylon that binds with some phenolic compounds can remove colour precursors and prevent enzymatic browning/pinking in whites. It may also clean up an imperfect wine’s odour/taste. Can remove anthocyanin colour, as in too-red blush wine. Eisenman (1991) elaborates in his paper on these fining materials.
Escolar et al. (1995) studied the decolourising effects of different quantities of activated carbon adsorbent on the colour of sherry samples (Fino white). Wine samples were treated with accurately weighed amounts of activated
carbon and mechanically shaken in a thermostatic bath at 250C for one hour to reach equilibrium, then centrifuged at 5000 rpm for five minutes and filtered with a 0.25-micrometer membrane filter. Data gathered showed that during the decolourisation process all the chromatic parameters varied, agreeing with the progressive disappearance of the colour yellow due to the chemical adsorption of the colouring compounds of the Fino wine, since the graphitised structure of the adsorbent shows affinity for polyphenolic compounds responsible for the colour. The colour change was measured through the use of chromatic parameters determined by the CIELAB 76 system, using transmittances taken at 1-nm intervals throughout the visible spectrum. The effect of the adsorbent on the colour of the wine has been expressed by equations that relate the values of the chromatic parameters with the quality of the added adsorbent. These equations show that the decolourisation process follows a Freundlich-type isotherm. He concluded, that, by using these equations, it is possible to predict the amount of carbon adsorbent necessary to attain a desired colour.
Pilone (1977) investigated methods determining the tartaric acid in wine and he does not specifically give attention to the decolourisation of wine. He does, however, use two methods of decolourisation before the tartaric acid concentration is determined. An activated carbon (Darco KB Activated Carbon, Atlas Chemical Industries Inc.) and a resin (20-50-mesh Bio-Beads-SM-2, Bio-Rad Laboratories) were used.
The POLYCLAR web page (www.iscorp.com, 2001) advertises two registered PVPP products to use in the wine industry, namely, Polyclar V, and Polyclar VT. These are used for stabilising wine to either prevent or remedy oxidation problems, caused either during the winemaking process, or due to high levels of grape rot and mould. In addition, Polyclar stabilisers are used to improve wine affected by over ageing or poor
storage conditions. Polyclar stabilisers are insoluble in wine and are completely removed in the lees by either filtration or rapid settling techniques. They are compatible with all fining agents and can also allow SO2 levels to be reduced.
Mennet and Nakayama (1969) tried casein, hide powder, nylon-66 powder, and Polyclar-AT (an insoluble polyvinylpirrolidine) as adsorbents. Adsorption isotherms in model systems and wine are evaluated for their adherence to the Freundlich isotherm. The effect of acidity on adsorption is investigated. The adsorption of the hydroxybenzoic acids by Polyclar-AT does not appear to be selective as determined by similar equilibrium constants for all the acids tested. The non-selectivity is due to the requirement for adsorption, which is the presence of one hydrogen-bonding donor. The acidity of a solute exerts a negative effect on its adsorption, although the effect is not strong enough to cause selective adsorption. 2.2.2 The chemistry of browning reactions
In this study it is necessary to have some understanding of how browning reactions occur, what their preferred environments are, as well as how these reactions could be impeded.
Researchers Kramling & Singleton (1965), Dutson & Orcutt (1984), Mayen et al. (1997) and Garza et al. (1999) collectively mention four basic pathways by which grape juice concentrate (GJC) can possibly undergo browning. They are: enzymatic oxidative browning, non-enzymatic oxidative browning, non-enzymatic browning (the Maillard reaction), and caramelisation.
We will group enzymatic- and non-enzymatic oxidative browning, since they are both oxidative, and explain each pathway in the next few paragraphs.
2.2.2.1 Oxidative browning
Oxidative browning is divided into two pathways, namely, enzymatic, and non-enzymatic. The main difference between these two pathways is the catalyst. In enzymatic oxidative browning the catalyst would be an enzyme, whereas the catalyst for non-enzymatic oxidative browning would be an acidic medium.
Enzymes are large proteins that act as catalysts for biological reactions. An enzyme is very specific in its action and will often catalyse only one specific reaction. Enzymes do not affect the equilibrium constant of a reaction and cannot bring about chemical changes that are otherwise unfavourable. Enzymes act only to lower the activation energy of a reaction, thereby making the reaction take place more rapidly (McMurry, 1996).
Some of the phenolics involved with oxidative browning reactions can be seen in Figure 2-2.
2.2.2.1.1 Enzymatic oxidative browning
This pathway involves enzymatic catalysts such as Polyphenol Oxidase (Phenolase, PPO), or Peroxidase to assist in the formation of o-quinones from phenolics (or antioxidants) in the juice, which polymerises to form brown-coloured polymers. Figure 2-3 indicates the pathway to brown polymers with PPO as catalyst.
Enzymes, of which the activity is not limited, can cause a serious reduction in the quality of the juice. Phenolase is relatively unstable to heat and can be inactivated by a mild heat treatment, whereas Peroxidase is much more heat resistant and needs higher temperatures for its destruction (Mayen et al., 1997).
O H O H O OH Caffeic Acid OH O O H OH OH OH Catechin O OH O O H OH OH OH Quercetin
FIGURE 2-2 A FEW ANTIOXIDANTS/PHENOLICS PARTICIPATING IN OXIDATIVE BROWNING REACTIONS
During the manufacturing process of the concentrate, the juice is subjected to mild heat in the concentration stages. It is subjected to somewhat higher heat during pasteurisation just before shipping. Not all enzymes involved in the browning reactions are inactivated by these "heat treatments". Their activity however, is significantly inhibited.
2.2.2.1.2 Non-enzymatic oxidative browning
Non-enzymatic oxidative browning, otherwise known as auto-oxidative browning, is catalysed by an acidic medium (See Figure 2-4).
As can be seen when Figure 2-3 and Figure 2-4 are compared, the products are quite similar, and both of them can be seen as a brown colour in the juice.
O O O H OH OH O OH O O H OH OH OH O H O O H OH OH OH OH O O H OH OH OH + H2O I II +I III PPO+O2
FIGURE 2-3 ENZYMATIC BROWNING OF CATECHIN, THE FORMATION OF A BROWN POLYMER OH OH OH OH OH O H O OH OH OH OH O H O CH+ H+ I II +I OH OH OH OH O H O OH OH OH OH O H O III
2.2.2.2 Non-enzymatic browning (The Maillard reaction)
The chemistry of the Maillard reaction is not fully understood in the context of this study. Although many studies have been conducted on the flavour formation of this sundry reaction, not much knowledge has been gathered concerning colour formation. The following paragraphs is a summary of the results from research work done by the following researchers: Resnik & Cherife (1979), Yeo & Shibamoto (1991), Arena et al. (2000), Petriella et al. (1985), Boston & Boyacioglu (1997), Beudo et al. (2001), Bozkurt et al. (1999), Litchfield et al. (1999) and Nagaraj & Monnier (1995).
The Maillard reaction consists of three stages (early stage, advanced stage and final stage) and depends upon factors such as pH, time, temperature, concentration of reactants and reaction type.
2.2.2.2.1 Early Stage
This stage involves the condensation of an amino acid with a reducing sugar to form Amadori or Heyns rearrangement products via an N-substituted glucosylamine, otherwise known as a Schiff base (Figure 2-5).
R1 H H OH O C C + H2N R R N R1 H H OH C C R NH HC R1 OH C H2C R1 O C R NH
Sugar Amine Schiff Base
1-amino-1-deoxy-2-ketose enol form
II
keto form I
FIGURE 2-5FORMATION OF SCHIFF BASE AND AMADORI COMPOUNDS
2.2.2.2.2 Advanced Stage
The degradation of the Amadori and Heyns rearrangement products can occur via four or five possible routes. We will only focus our attention on one of them, namely the pathway to the formation of 5-hydroxymethylfurfural (HMF), since HMF is a known precursor in
browning reactions. Figure 2-6 indicates the pathway to the formation of HMF. R1 OH R OH OH HC N H CH C HC OH R1 O HC O HC C CH2 R1 O CH O HC C CH H O O H O -OH --amine +H2O -H2O -H2O + amine melanoidins 5-hydroxymethylfurfural H N+ R OH R1 HC OH HC HC CH2
FIGURE 2-6THE MAILLARD REACTION PATHWAY TO HMF AND MELANOIDINS
2.2.2.2.3 Final Stage
The final stage of the Maillard reaction is characterised by the formation of brown nitrogenous polymers and co-polymers. As mentioned earlier very little is known about the chemical nature of these products. What is known, however, is that these colour compounds can be grouped into two general classes - low molecular weight colour compounds, which comprises of two to four linked rings, and the melanoidins, which have much higher molecular weights. Colour development increases with increasing temperature, time of heating, and decreasing pH.
2.2.2.3 Caramelisation
Only a slight overview of caramelisation is given here since this pathway to browning has no significant contribution in the process in this study. Caramelisation is also a non-enzymatic browning reaction, but one that involves the heat-induced decomposition of sugars, normally
monosacharides. They undergo initial enolisation, known as the Lobry de Bruyn-A. van Eckenstein rearrangement, and progress to subsequent complex reactions, such as dehydration, dicarboxylic cleaving and aldol condensation. The reaction generally releases H+, thus the pH of the solution decreases with time (Lee & Lee, 1997)
2.2.2.4 Pathways to the formation of HMF
In the juice and juice concentrate studied, the concentration of 5-hydroxymethylfurfural (HMF) is of great importance. Gomis et al. (1991) has shown that HMF can be used as a determinative factor for the browning potential of juice concentrate. HMF is, of course, a precursor to browning as can be seen in Figure 2-6 and it is important to gain knowledge on how this compound is formed. From literature (Graza et al., 1999, and Feather, 1982) two pathways are observed.
The first is as shown in Figure 2-6, where the formation of HMF is amine assisted (the Maillard reaction). The second pathway is an acid catalysed reaction (non-enzymatic oxidative browning). According to Feather (1982) there is a great deal of similarity between the dehydration of a sugar in acidic solution and in the presence of amines. This similarity can clearly be seen when the reactions in Figure 2-6 and Figure 2-7 are compared. The formation of a 2-Furaldehyde (such as HMF) via acid catalysis is shown in Figure 2-7. Further reaction with an amine will yield a melanoidin, which is visible as a brown colour in high concentrations. According to Feather (1982) the reducing sugars decompose at much milder conditions in the presence of amino groups (amines) than for the acid catalysed reaction.
2.2.2.5 Preferred environments of browning pathways
The preferred environment of the various pathways of browning reactions is summarised in Table 2-1.
OH R1 O HC O HC C CH2 R1 O CH O HC C CH -H2O -H2O -H2O 2-furaldehyde OH R1 OH OH OH HC HC C HC O OH R1 HC OH HC C CH R H O O
FIGURE 2-7ACID CATALYSED SUGAR DEHYDRATION (THE FORMATION OF HMF) TABLE 2-1 A SUMMARY OF THE PREFERRED ENVIRONMENTS OF THE BROWNING REACTIONS
Browning Reaction Preferred Environments
Enzymatic Oxidative browning Mild temperatures, mild acidic environment
Non-Enzymatic Oxidative Browning Acidic environment, higher temperatures The Maillard Reaction Acidic environment, high temperature Caramelisation Acidic environment, very high
2.2.2.6 Grouping of some of the phenolics
Table 2-2 contains some of the major groups of grape juice phenolics and examples of them are given.
TABLE 2-2EXAMPLES OF GRAPE JUICE PHENOLICS
Major groups Members Example
Hydroxybenzoic Acids Gallic Acid Protocatechuic Acid Vanillic Acid O H O H O H O OH GALLIC ACID Hydroxycinnamic Acids Caffeic Acid Chlorogenic Acid Ferulic Acid p-Coumaric Acid O H O H O OH Caffeic Acid Flavan-3-ols Catechin Epicatechin CATECHIN OH O O H O H OH OH Others Quercetin Resveratrol O OH O O H OH OH OH Quercetin Furfurals HMF 2-Furaldehyde H O O H O 5-Hydroxymethylfurfural
The examples given in Table 2-2 were chosen as initial representatives of the phenolics and furfural involved in this study. Added to these was Resveratrol.
2.2.3 Applications and limitations of adsorption products
This section contains information on the applications and limitations of some adsorption products mentioned in the previous sections. Other topics
mentioned are: operating parameters, regeneration/reactivation/ disposal determination, and operating and maintenance costs. Also some technical guidelines will be given on how to decide when alternative adsorption media may be selected in lieu of activated carbon.
2.2.3.1 Carbon adsorption
2.2.3.1.1 Applications
Some typical rules of thumb for types of compounds that are amenable to carbon adsorption are as follows:
• Larger molecules adsorb better than smaller molecules; • Non-polar molecules adsorb better than polar molecules;
• Non-soluble or slightly soluble molecules adsorb better than highly soluble molecules;
• Based on the polarity or solubility (or both) of the molecule being adsorbed, pH may have an influence on the extent of adsorption; • Temperature increases the rate of diffusion through the liquid to the
adsorption sites, but since the adsorption process is exothermic, increases in temperature may reduce the degree of adsorption. This temperature effect is negligible in water treatment applications and ambient vapour phase applications.
2.2.3.1.2 Chemicals Adsorbed
The following are examples of chemicals adsorbed:
• Alcohols are poorly adsorbed since they are very soluble and highly polar;
• Aldehydes are highly polar, and as molecular weight increases, the polarity decreases, and adsorbability increases;
• Amines are similar in structure to ammonia (NH3), except the
polarity and solubility. Typical amines in grape juice would be proteins and amino acids. It has been shown that these compounds decrease the adsorption potential of activated carbon;
• Chlorinated aromatics, and chlorinated aliphatics are low-polarity and low-solubility compounds, which make them generally quite adsorbent;
• Glycols are water-soluble and not very adsorbent. Glucose and fructose can be included in this category. Both are naturally occurring sugars in grape juice;
• Higher molecular weight organic compounds will generally be more adsorbant owing to adsorptive attraction relative to size.
2.2.3.1.3 Operating Parameters
The following parameters should be taken into consideration when designing an adsorption process:
• Contact time - the time the activated carbon will have to be in contact with the liquid to remove the undesired compounds efficiently;
• Adsorbent volume - the volume of activated carbon needed to remove undesired compounds to a specified concentration in the contact time determined;
• Equipment needed - including pumps, mixing facilities, contact tanks, and filtration facilities;
• Regeneration, reactivation and disposal.
2.2.3.1.4 Regeneration, Reactivation, and Disposal
This section is specifically focused on granular activated carbon with some attention given to powder activated carbon (as used on the concentrate manufacturing plant).
As contaminants are adsorbed, the carbon’s adsorptive capacity gradually decreases. When the carbon’s adsorptive capacity is reached, it is considered “spent,” and it must be regenerated, reactivated, or disposed of. Regeneration usually involves removing the adsorbed contaminants from the carbon using temperatures or processes that drive the contaminants from the carbon but that do not destroy the contaminants or the activated carbon. A common regeneration process introduces steam into the spent carbon bed, volatilising the contaminants and restoring the carbon’s capacity to what is called its “working capacity.” Steam regeneration does not completely remove adsorbed contaminants. Another common process uses a hot inert gas, such as nitrogen, to remove the contaminants. The stripped volatiles are compressed, and recovered as liquid in a condenser. A third process is pressure swing adsorption. Pressure swing adsorption uses the fact that adsorption capacity is directly proportional to the partial pressure of the contaminants in the surrounding environment. The contaminants are adsorbed at a high pressure (providing higher partial pressure of the contaminant to be adsorbed), and then desorbed at a lower pressure where the capacity is reduced. These regeneration processes are usually run on-site and inside the adsorption vessel. All regeneration processes produce a waste stream that contains the desorbed contaminants. For example, steam regeneration produces a mixture of water and organics from the condensed desorbed vapour. This is not at all practical for use in powdered activated carbon regeneration. Powdered activated carbon is firstly filtered from the liquid media. This is done by means of a diatomaceous earth filtering aid. To separate these two is virtually impossible. Regeneration will not be an attainable task.
Other than thermal reactivation at elevated temperatures, regeneration techniques will result in some contaminants remaining adsorbed and unaltered within the carbon particle. These contaminants will be occupying
“high energy adsorption pores, or sites,” and lower temperature regenerants (< 260°C) or capacity corrections will not be able to provide sufficient energy to reverse the adsorptive force. Carbon having these residual contaminants remaining in the high-energy adsorption sites will likely have much shorter runs before breakthrough. Protein and amino acids are known to contaminate activated carbon by occupying these high-energy adsorption pores.
Spent carbon reactivation off-site, involves removing the adsorbed contaminants from the spent activated carbon in a process that is a modification of the one that initially activated the carbon. The contaminants are desorbed and destroyed in the high temperature (typically in excess of 800°C), pyrolising atmosphere of the reactivation furnace. Several types of furnaces are available, such as rotary kilns and multiple hearths. The furnaces can be heated by a fuel such as natural gas or fuel oil or by electricity. Off-site carbon reactivation manufacturers reactivate spent carbon in large capacity (5 to 60 tons/day) furnaces. While furnaces of this capacity are not typically cost effective for a single hazardous waste site, smaller furnaces that may prove cost effective are available for on-site use from a few manufacturers. Reactivation furnaces only produce reactivated carbon, air emissions, and some carbon fines. No organic wastes are produced. Again, with the filtering aid and powdered activated carbon, this option is not extremely viable, unless another, inexpensive method of filtration exists.
2.2.3.1.5 Criteria for Determining When to Use On-site Regeneration, Reactivation or Off-site Reactivation, or Disposal
• On-site reactivation requires space and utility support for the equipment. It also usually requires an air pollution permit for the furnace afterburner. If the site cannot provide the land or utility support, or if obtaining the required permit is not practical, the
spent carbon must rather be regenerated on-site (if possible) or reactivated off-site;
• At some sites, the availability or turn-around times for off-site carbon re-supply may be impractical. In these situations, on-site regeneration or reactivation will be required or the site can provide sufficient storage for both fresh and spent carbon to eliminate the constraint of response time by outside suppliers;
• Studies indicate that on-site thermal reactivation is not economical if carbon usage is less than 500 to 2000 lb/day (227 to 909 kg/day). Other studies have found that carbon reactivation unit cost rises rapidly if carbon usage is less than 5000 to 6000 lb/day (2272 to 2727 kg/day);
• When carbon is regenerated on-site, some contaminants may not be desorbed. For example, GAC containing organic contaminants with high boiling points may need to be reactivated instead of regenerated;
• There are several cases where regeneration or reactivation of the spent carbon will not be feasible or will be prohibitively expensive. In these cases, the spent carbon must be disposed of.
2.2.3.2 Non-Carbon Adsorption
Modified clay, polymeric adsorbents, and zeolite molecular sieves are also currently used as adsorbents in a variety of applications. Some of these adsorption media are used primarily as pre-treatment for activated carbon. For example, these media may be used to remove compounds that may, through physical or chemical interactions, degrade the effectiveness of the activated carbon. As mentioned earlier amines such as proteins and amino acids can significantly degrade the effectiveness of activated carbon during the decolourisation of grape juice, due to fouling.
